Date Published: June 16, 2012
Publisher: Impact Journals LLC
Author(s): Alejandro Vazquez-Martin, Sílvia Cufí, Bruna Corominas-Faja, Cristina Oliveras-Ferraros, Luciano Vellon, Javier A. Menendez.
Recent studies have suggested a pivotal role for autophagy in stem cell maintenance and differentiation. Reprogramming of somatic cells to induced pluripotent stem cells (iPSCs) has been also suggested to bio-energetically take advantage of mitochondrial autophagy (mitophagy). We have preliminary addressed how mitophagy might play a role in the regulation of induced pluripotency using mdivi-1 (for mitochondrial division inhibitor), a highly efficacious small molecule that selectively inhibits the self-assembly of DRP1, a member of the dynamin family of large GTPases that mediates mitochondrial fission. At mdivi-1 concentrations that rapidly induced the formation of mitochondrial net-like or collapsed perinuclear mitochondrial structures, we observed that the reprogramming efficiency of mouse embryonic fibroblasts transduced with the Yamanaka three-factor cocktail (OCT4, KLF4, and SOX2) is drastically reduced by more than 95%. Treatment of MEFs with mdivi-1 at the early stages of reprogramming before the appearance of iPSC colonies was sufficient to completely inhibit somatic cell reprogramming. Therefore, the observed effects on reprogramming efficiencies were due likely to the inhibition of the process of reprogramming itself and not to an impairment of iPSC colony survival or growth. Moreover, the typical morphology of established iPSC colonies with positive alkaline phosphatase staining was negatively affected by mdivi-1 exposure. In the presence of mdivi-1, the colony morphology of the iPSCs was lost, and they somewhat resembled fibroblasts. The alkaline phosphatase staining was also significantly reduced, a finding that is indicative of differentiation. Our current findings provide new insight into how mitochondrial division is integrated into the reprogramming factors-driven transcriptional network that specifies the unique pluripotency of stem cells.
Because stem cells need to protect their genome from damage to maintain both the progenitor pool and their self-renewal capacity  and because intracellular ROS levels influence the long-term self-renewal capacity of HSCs [13-15], the above-mentioned studies strongly suggest that mitophagy protects the genome due to its ability to clear mitochondria as a source of ROS; therefore, mitophagy may help stem cells to maintain their self-renewal and pluripotent capacities [16-18]. However, it remains to be elucidated whether mitophagy is mechanistically linked to the acquisition of pluripotency. Recent studies have demonstrated that undifferentiated pluripotent stem cells display lower levels of mitochondrial mass and oxidative phosphorylation and that they preferentially use non-oxidative glycolysis as a major source of energy. Folmes and colleagues  confirmed that the stemness factor-mediated reprogramming of somatic cells into induced pluripotent stem cells (iPSCs) remarkably reverts mitochondrial networks into cristae-poor structures. Second, as has been previously shown by Prigione & Adjaye , the functional metamorphosis of somatic oxidative phosphorylation into acquired pluripotent glycolytic metabolism corresponds to an embryonic-like original pattern . Thus, somatic mitochondria within human iPSCs suffer a reversion to an immature embryonic stem cell (ESC)-like state with respect to organelle morphology, distribution, and function, suggesting that the mitochondrial/oxidative stress pathway is actively modulated during cellular reprogramming to induce a rejuvenated state capable of escaping cellular senescence . Indeed, Folmes’s metaboproteomic studies demonstrated that cell fate during reprogramming is determined by the upregulation of glycolytic enzymes and the downregulation of electron transport chain complex I subunits. Temporal sampling demonstrated glycolytic gene potentiation prior to the induction of pluripotent markers; accordingly, stimulating glycolysis promotes reprogramming, and inhibiting glycolytic enzyme activity blunts reprogramming efficiency [19, 22]. Panopoulos and colleagues  have recently confirmed that a bioenergetic shift from somatic oxidative mitochondria toward an alternative ATP-generating glycolytic phenotype maximizes the efficiency of somatic reprogramming to pluripotency. In their hands, somatic cells that demonstrated oxidative:glycolytic energy production ratios closer to pluripotent cells reprogrammed more quickly and efficiently. Altogether, these studies strongly suggest that changes in metabolism may play a role in enabling the reprogramming process to occur rather than simply being a consequence of acquiring a pluripotent state.
Two distinct dynamin-related GTPases (DRPs), which function via self-assembly to regulate membrane dynamics in a variety of cellular events, are required for mitochondrial fusion [39, 40]. MFN1/2/Fzo1 (human/yeast nomenclature) and OPA1/Mgm1 drive outer and inner mitochondrial membrane fusion, respectively. A single DRP, DRP1/Dnm1, is required for mitochondrial fission [40, 41]. DRP1 is assembled from the cytosol onto mitochondria at focal sites of division , forming spiral chains around membrane constriction sites . DRP1 self-assembly facilitates GTP hydrolysis and thereby organelle fission. In mammalian cells, when mitochondrial division is retarded by the expression of dominant-negative DRP1 or by RNAi of mitochondrial division proteins, tubular mitochondria become progressively more interconnected to form net-like structures and also collapse into degenerate perinuclear structures. However, overexpression of wild-type DRP1 does not lead to mitochondria fragmentation, suggesting that a simple alteration of DRP1 levels could not alter mitochondrial fission. Regulation of DRP1 properties, such as mitochondrial translocation, higher order assembly or GTPase activity, is critical [36, 39]. Here, we used a small molecular inhibitor of DRP1 to probe the mechanistic role that mitochondrial division plays in both the acquisition and the maintenance of pluripotency. We employed mdivi-1 (for mitochondrial division inhibitor), an inhibitor of mitochondrial division identified by Cassidy-Stone and colleagues using yeast screens of chemical libraries . Because it is thought that mitochondrial fission is related to the progression of mitophagy, the inhibition of mitochondrial fission by mdivi-1, a specific inhibitor of DRP1-GTPase, has been shown to compromise mitophagy . The addition of mdivi-1 to mammalian cells in culture has been shown to cause a rapid and reversible formation of mitochondrial net-like and degenerate perinuclear structures, consistent with attenuation in mitochondrial division [31, 33]. Indeed, depletion of DRP1 by RNAi causes the formation of net-like or collapsed perinuclear mitochondrial structures in mammalian cells, and treatment of these cells with mdivi-1 does not produce any additional changes to mitochondrial morphology, thus substantiating that DRP1 is the specific target of mdivi-1 in mammalian cells . Our first step in determining the function DRP1-regulated mitochondrial dynamics was to confirm the effects of mdivi-1 on mitochondrial morphology. Using DsRed-Mito to label mitochondria, control MEFs displayed primary tubular and long mitochondria (Fig. 1, top). mdivi-1 treatment caused the formation of net-like mitochondria, as expected from its ability to directly attenuate mitochondrial division (Fig. 1, bottom).
To address the functional effects of a mitochondrial fission deficit imposed by DRP1-GTPase inhibition on iPSC generation, we performed comparison experiments using the three-factor (i.e., OCT4, SOX2, and KLF4) induction protocol in early-passage mouse embryonic fibroblasts (MEFs). MEFs were first transduced with individual lentiviruses containing OCT4, SOX2, and KLF4 at a 1:1:1 ratio on day 0. The transduction was repeated every 12 h for 2 days using the same batch of all three lentiviruses. On day three after the first transduction, the culture medium was switched to human embryonic stem (hES) cell growth medium with or without two different concentrations of mdivi-1. We used mdivi-1 at 10 and 50 μmol/L, the range of mdivi-1 concentrations required to observe either net-like or collapsed/degenerate perinuclear mitochondrial structures [31, Fig. 1, bottom]; the ES medium with or without mdivi-1 was renewed every two days. From days 10-12, clearly recognizable, tightly packed colonies similar to hES cells appeared in the mdivi-1-free control cultures. We then combined the observations of ES cell-like morphological changes (e.g., defined boundaries and high nucleus-to-cytoplasm ratio within individual cells) with alkaline phosphatase (AP) staining, a commonly used pluripotency marker, to quantify bona fide iPSC colonies on day 14 post-viral transduction.
Alkaline phosphatase (AP) is a universal pluripotency marker for all types of pluripotent stem cells, including embryonic stem cells, embryonic germ cells and iPSCs. Indeed, AP staining is widely used to identify emerging pluripotent colonies during the process of somatic reprogramming. We employed AP staining and observed the morphological changes of iPSCs to evaluate whether mitochondrial division is required for the maintenance of the undifferentiated state of iPSCs. At day 20 post-transduction, the established iPSCs were selected and passaged onto pre-seeded MEF feeder cells. The iPSCs were then exposed to two different concentrations of mdivi-1 for 5 days. Strong and uniform AP staining was detected in the untreated iPSC control colonies, demonstrating one of the properties attributed to pluripotent cells (Fig. 3). Remarkably, the typical morphology of established iPSC colonies with AP+ staining was significantly affected by mdivi-1 exposure. On the one hand, the colony morphology of the iPSC colonies (i.e., dense round cells with a well-defined edge) was lost, and they rather resembled fibroblast-like differentiated cells (Fig. 3). On the other hand, the AP staining was drastically reduced, which is indicative of differentiation. Thus, the prevention of mitochondrial division imposed by mdivi-1-inhibited DRP1 apparently led to differentiation and consequently disrupted the self-renewal of iPSCs.
Mitochondria certainly should play a role in the metabolic shift that enables somatic reprogramming to stemness because the physiology of mitochondria is inextricably linked to energy metabolism . Specifically, mitochondrial structure and function have been suggested to be indicators of stem cell competence because low mitochondrial activity and relatively under-developed mitochondrial networks have been confirmed to be common features of stemness [43-48]. Vessoni and colleagues  hypothesized that autophagy could play an important role in mediating the remodeling of differentiated cells to a pluripotent state during the generation of iPSCs. Mitophagy would promote mitochondrial degradation during iPSC generation, allowing differentiated cells to reduce the amount of this organelle to ESC-like levels. To test a “metabolic state hypothesis” that links the mitochondrial state and cellular bioenergetics to the state of differentiation, Vessoni and colleagues  suggested that an increase in the number of developed mitochondria and the mitochondrial mass in iPSCs generated from autophagy-deficient cells (ATG7−/−) would argue for a pivotal role for autophagy during reprogramming. In the same way, the generation of iPSCs from differentiated cells might also be positively influenced by autophagy modulation. Because mitochondrial fission is a mediator of mitochondrial turnover (i.e., mitochondrial fission followed by selective fusion segregates dysfunctional mitochondria and permits their removal by autophagy) and because inhibiting mitochondrial fission results in the specific inhibition of mitochondrial autophagy before the phagophore is assembled [50, 51], we recently envisioned that pharmacological perturbation of mitochondrial dynamics before and after iPSC generation may illuminate mitophagy as a pivotal mechanism driving somatic reprogramming to stemness. Our current findings provide new insight into how mitochondria division is integrated into the reprogramming factor-driven transcriptional network that specifies the unique pluripotency of stem cells. Our data strongly suggest for the first time that the occurrence of mitophagy may be involved in the selective turnover of mitochondria prior to and during the reprogramming of somatic cells to iPSCs. In light of recent studies suggesting that changes in metabolism may play a role in enabling the reprogramming process to occur, instead of being a consequence of acquiring a pluripotent state, our data confirm a causal correlation between the bioenergetic state of somatic cells and their reprogramming efficiency. Future studies should elucidate whether the ability of mitophagy to directly shift the oxidative:glycolytic production ratios closer to those of pluripotent cells (i.e., somatic cells primarily utilize mitochondrial oxidation for proliferation, whereas pluripotent cells favor glycolysis) can molecularly explain the impact of mitochondria fusion/fission dynamics on the acquisition and maintenance of stem cell pluripotency.